Control of FLIPL expression and TRAIL resistance by the ...file2.selleckchem.com/citations/getinib-U0126-20120343.pdfthe extracellular signal-regulated kinase1/2 pathway in breast

Control of FLIPL expression and TRAIL resistance bythe extracellular signal-regulated kinase1/2 pathwayin breast epithelial cells

R Yerbes1, A Lopez-Rivas*,1, MJ Reginato2 and C Palacios*,1

Increased activation of the epidermal growth factor receptor (EGFR) is frequently observed in tumors, and inhibition of thesignaling pathways originated in the EGFR normally renders tumor cells more sensitive to apoptotic stimuli. However, we showthat inhibition of EGFR signaling in non-transformed breast epithelial cells by EGF deprivation or gefitinib, an inhibitor of EGFRtyrosine kinase, causes the upregulation of the long isoform of caspase-8 inhibitor FLICE-inhibitory protein (FLIPL) and makesthese cells more resistant to the tumor necrosis factor-related apoptosis-inducing ligand (TRAIL). We demonstrate that theextracellular signal-regulated kinase (ERK)1/2 pathway plays a pivotal role in the regulation of FLIPL levels and sensitivity toTRAIL-induced apoptosis by EGF. Upregulation of FLIPL upon EGF deprivation correlates with a decrease in c-Myc levels andc-Myc knockdown by siRNA induces FLIPL expression. FLIPL upregulation and resistance to TRAIL in EGF-deprived cells arereversed following activation of an estrogen activatable form of c-Myc (c-Myc-ER). Finally, constitutive activation of the ERK1/2pathway in HER2/ERBB2-transformed cells prevents EGF deprivation-induced FLIPL upregulation and TRAIL resistance.Collectively, our results suggest that a regulated ERK1/2 pathway is crucial to control FLIPL levels and sensitivity to TRAIL innon-transformed cells, and this mechanism may explain the increased sensitivity of tumor cells to TRAIL, in which the ERK1/2pathway is frequently deregulated.Cell Death and Differentiation (2012) 19, 1908–1916; doi:10.1038/cdd.2012.78; published online 22 June 2012

Tumor necrosis factor (TNF)-related apoptosis-inducingligand (TRAIL) is a ligand of the TNF family capable ofinducing apoptosis in a wide variety of cancer cells uponbinding to pro-apoptotic receptors, but with no effect in themajority of normal human cells tested.1 This unique char-acteristic of TRAIL is currently being exploited as a potentialantitumor therapy.2 Activation of TRAIL receptors (TRAIL-Rs)leads to the formation of the death-inducing signaling complex(DISC), which includes the receptor itself, the adaptermolecule Fas-associated death domain (FADD) and procas-pase-8.3,4 Processing and activation of caspase-8 at the DISCleads to a cascade of apoptotic events resulting in cell death.At the DISC level, the apoptotic signal may be inhibited bycellular FLICE-inhibitory protein (cFLIP), the homolog of viralFLIP (vFLIP) in vertebrate cells.5 In most cells, cFLIP existsas two alternative spliced isoforms: FLIPL, a homologof caspase-8 that lacks critical amino acids for proteolyticcaspase activity, and cFLIPS, consisting only of two deatheffector domains (DED).5 Although the role of FLIPL as aninhibitor of apoptotic signaling has been controversial,6,7

recent data indicate that heterodimerization of FLIPL with

procaspase-8 induces limited caspase-8 activation, prevent-ing further transduction of the apoptotic signal.8 FLIPexpression fluctuates in a cell-type-specific manner and inresponse to various stimuli, transcriptionally through theNF-kB pathway,9 and at the protein level via altered rates ofproteasomal degradation,10 which makes it a versatileinhibitor of apoptotic responses mediated by death receptors.

Binding of epidermal growth factor (EGF) to its monomericreceptor (EGFR) activates homo/heterodimerization and self-phosphorylation on tyrosine residues, which are used bydocking proteins to engage molecules involved in differentsignaling pathways.11 Activation of the Ras-dependent extra-cellular signal-regulated kinase (ERK)1/2 mitogen-activatedprotein kinase (MAPK) pathway by EGF in normal cells isrequired for efficient G1- to S-phase transition and for thecontrol of cell proliferation.12 Upon activation, ERK1/2translocates to the nucleus and phosphorylates the ternarycomplex factor, which in turn induces the transcription ofimmediate-early genes like c-Fos and c-Myc.13,14 ProlongedERK activation results in c-fos and c-myc phosphorylation,stabilizing these transcription factors and thereby promoting

1Centro Andaluz de Biologıa Molecular y Medicina Regenerativa (CABIMER), Consejo Superior de Investigaciones Cientıficas, Sevilla, Spain and 2Department ofBiochemistry and Molecular Biology, Drexel University College of Medicine, Philadelphia, PA, USA*Corresponding authors: A Lopez-Rivas or C Palacios, Centro Andaluz de Biologıa Molecular y Medicina Regenerativa (CABIMER), Consejo Superior deInvestigaciones Cientıficas, Avenida Americo Vespucio s/n, 41092 Sevilla, Spain. Tel: +34 954 467997 or +34 954 467934; Fax: +34 954 461664;E-mail: [email protected] (AL-R); E-mail: [email protected] (CP)

Received 11.1.12; revised 24.4.12; accepted 15.5.12; Edited by M Deshmuck; published online 22.6.12

Keywords: ERK1/2; c-myc; FLIPL; TRAIL; apoptosisAbbreviations: EGF, epidermal growth factor; EGFR, epidermal growth factor receptor; ERK, extracellular signal-regulated kinase; 4HT, 4-hydroxytamoxifen; MAPK,mitogen-activated protein kinase; TRAIL, tumor necrosis factor-related apoptosis-inducing ligand; TRAIL-R, TRAIL receptor; TNF, tumor necrosis factor;FADD, Fas-associated death domain; FLIP, FLICE-inhibitory protein; DISC, death-inducing signaling complex; MEK1, MAPK/ERK kinase 1; PI3K, phosphatidyl inositol3-kinase

Cell Death and Differentiation (2012) 19, 1908–1916& 2012 Macmillan Publishers Limited All rights reserved 1350-9047/12

www.nature.com/cdd

http://dx.doi.org/10.1038/cdd.2012.78

mailto:[email protected]

mailto:[email protected]

http://www.nature.com/CDD

activation of the G1-specific cyclin-dependent kinases andcell cycle entry.15,16

In many human cancers, including breast carcinoma,EGFR overexpression is correlated with cellular proliferation,angiogenesis and tumor growth, leading to invasiveness andmetastasis,17 thus providing a potential target for therapy.18

In several tumor cell types, activation of the EGFR inducesresistance to different apoptotic stimuli, including TRAIL,19

due to activation of the phosphatidyl inositol 3-kinase (PI3K)/AKT pathway. However, despite all the available evidences intumor cells, the regulation of TRAIL sensitivity by EGFRsignaling in non-tumor cells remains to be investigated.We report here that EGFR activation downregulates cFLIPexpression and enhances the sensitivity of non-transformedhuman breast epithelial cells to TRAIL-induced apoptosis.Our results also indicate that cFLIP levels and TRAILsensitivity are both controlled by the EGFR-mediated regula-tion of the ERK1/2 pathway and c-myc expression. In addition,we show that deregulated ERK1/2 activation in human breastepithelial cells transformed with the oncogene Her2/ERBB2prevents EGF deprivation-induced FLIPL upregulation andTRAIL resistance, underlining the relevance of this pathwayin controlling TRAIL-induced apoptosis.

Results

EGF controls the sensitivity of human breast epithelialcells to TRAIL-promoted apoptosis at the apical level inTRAIL-R signaling. In the past years, several compoundstargeting the EGFR have been developed to treat advancedcancers.18 Moreover, inhibition of EGFR signaling syner-gizes with TRAIL in the induction of apoptosis in tumorcells.20 However, cross-talk between EGFR and TRAIL-R

signaling has not been investigated in non-tumor cells.To address this issue, we determined the effect of EGFdeprivation on the apoptotic response to TRAIL in non-tumorbreast epithelial cell lines. Results depicted in Figure 1a andSupplementary Figure S1A and S1C demonstrate thatpreventing EGFR signaling by EGF deprivation inducedresistance to TRAIL. These results we confirmed withgefitinib, an inhibitor of the EGFR tyrosine kinase and EGFRsignaling.21 Incubation of MCF10A cells in EGF-containingmedium (CM) with gefitinib completely inhibited TRAIL-induced apoptosis (Supplementary Figure S2A), furtherindicating that EGFR activation facilitates TRAIL-inducedapoptosis in non-tumor cells.

Activation of pro-apoptotic receptors by TRAIL inducesapical procaspase-8 cleavage to generate the small subunit,p12, and the p43/41 intermediate fragment, which is subse-quently processed to produce the large catalytically active p18subunit.4 We determined the processing of procaspase-8 inMCF10A cells treated with TRAIL following culture of cellsin the presence or absence of EGF. Procaspase-8 cleavage togenerate its 43–41 kDa intermediate fragments, and thesubsequent generation of p18 subunit upon TRAIL-R activa-tion were markedly inhibited in MCF10A cells deprived of EGF(Figure 1b).

Upregulation of FLIPL upon EGF deprivation confersresistance to TRAIL. Expression levels of DISC proteinssuch as TRAIL-R2, FADD or procaspase-8 did not changefollowing EGF deprivation in MCF10A cells (Figures 1c andd), indicating that a reduction of pro-apoptotic DISCcomponents was not the mechanism underlying the dimin-ished caspase-8 activation by TRAIL in EGF-deprived cells.Interestingly, EGF withdrawal caused the upregulation of the

Ap

op

toti

c ce

lls (

%)

0

20

40

60

80

100

EGF + -

ControlTRAIL

*

Caspase-8

p18

GAPDH

p55/53p43/41

+ - + - + - + -

15 30 60

EGF

Time (min)

+ TRAIL

Procaspase-8

FADDGAPDH

EGF + -

TRAIL-R1

Fluorescence

TRAIL-R2

Cel

ls

FLIPL

FLIPS

TRAIL-R2

Figure 1 EGF deprivation induces resistance to TRAIL-induced apoptosis at the apical level. (a) MCF10A cells grown in the presence or absence of 20 ng/ml EGF during48 h were treated with or without TRAIL (500 ng/ml) for 6 h. Apoptosis was then measured as described in Materials and Methods. Error bars represent S.D. from threeindependent experiments. *Po0.05. (b) MCF10A cells were grown in the presence or absence of EGF for 48 h before adding TRAIL (1mg/ml). Cells were incubated withTRAIL for the indicated times and activation of caspase-8 was assessed by western blotting. In the TRAIL-untreated samples ,FLIPL, FLIPS, TRAIL-R2, procaspase-8 andFADD levels were also determined. (c) Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used as a protein loading control. Results are representative of twoindependent experiments. (d) Cells incubated with (�� ) or without EGF (—) for 48 h were harvested and cell surface expression of TRAIL-R1 and TRAIL-R2 receptorsassessed by flow cytometry as described in Materials and Methods. Cells incubated with FITC-labeled secondary antibody alone were used as a control for backgroundfluorescence of cells cultured in the presence (.....) or absence (- - - -) of EGF. Results are representative of two independent experiments

ERK1/2 controls FLIPL levels and TRAIL resistanceR Yerbes et al

1909

Cell Death and Differentiation

long form of cFLIP (FLIPL) in non-transformed breastepithelial cells (Figure 1c and Supplementary Figures S1Band S1D), suggesting a possible role of this inhibitor in theenhanced resistance to TRAIL of cells deprived of EGF.Furthermore, inhibition of EGFR signaling with gefinitibmarkedly upregulated FLIPL expression in MCF10A cells(Supplementary Figure S2B), confirming that EGFR activa-tion by EGF is involved in the regulation of FLIPL levels innon-transformed breast epithelial cells. In contrast, depriva-tion of either insulin or hydrocortisone, two medium additivesfor optimal growth of MCF10A cells,22 did not upregulateFLIPL expression (Supplementary Figure S2C).

Time-course analysis of FLIPL levels following EGFdeprivation (Figure 2a, left panel) indicated that FLIPL, butnot FLIPS, markedly accumulated in MCF10A cells between40 and 72 h after EGF removal. FLIPL levels slightly increasedafter 72 h incubation in the presence of EGF (Figure 2a, rightpanel), which was associated to a reduction in EGFRsignaling as determined by the inhibition of ERK1/2 phos-phorylation, most likely reflecting the impact of cell density on

EGFR signaling as reported recently.23 We next determinedwhether the upregulation of FLIPL protein seen in EGF-deprived cells was also observed at the mRNA level.RT-qPCR analysis of mRNA levels demonstrated that EGFdeprivation induced a marked upregulation of FLIPL mRNAstarting at 40 h after EGF withdrawal (Figure 2b). Strikingly, astrong correlation was observed between the decrease inEGFR signaling (ERK1/2 phosphorylation) and the increasedexpression of FLIPL and resistance of MCF10A cells to TRAILfollowing EGF deprivation (Figures 2a and c).

As FLIPL levels are important determinants of the sensitivityof breast tumor and non-tumor epithelial cells to TRAIL-induced apoptosis,24,25 we then assessed whether silencingFLIPL expression could sensitize EGF-deprived cells toTRAIL. Interestingly, FLIPL knockdown (Figure 3a, upperpanel) was sufficient to prevent the induction of TRAILresistance in EGF-deprived cells (Figure 3a, lower panel),indicating the importance of FLIPL accumulation in theacquired resistance to TRAIL after EGF withdrawal. Further-more, in cells stably overexpressing FLIPL (Figure 3b, upperpanel), TRAIL-induced apoptosis was clearly inhibited inEGF-CM (Figure 3b, lower panel). Our results supported animportant role of EGF-regulated FLIPL levels in the resistanceof non-transformed breast epithelial cells to TRAIL-inducedapoptosis.

Role of the ERK1/2 pathway in the control of FLIPL levelsand TRAIL resistance in breast epithelial cells. UponEGFR activation by EGF, the MAPK/ERK and the PI3K/Aktpathways, among others, are activated to produce aphysiological outcome.11 To investigate the role of these

0

2

4

6

8FLIPL

FLIPS

0 15 24 40 48 72

15 24 40 48 72 Time (h)

Time without EGF (h)

mR

NA

fo

ld e

xpre

ssio

n

0

20

40

60

80

100

+ - + - + - + - + - EGF

*** *****

Ap

op

toti

c ce

lls (

%)

FLIPL

FLIPS

Tubulin

0 15 24 40 48 72

Longer Exposure

pERK

FLIPL

Time without EGF (h)

ERK1/2

0 72

+ + - EGF

Time (h)

FLIPL

pERK 1/2

ERK 1/2

GAPDH

NoTRAILTRAIL

Figure 2 Upregulation of FLIPL upon EGF deprivation correlates withresistance to TRAIL. Analysis by western blotting (a) or quantitative PCR (b) ofFLIPL and FLIPS levels in MCF10A cells deprived of EGF for the indicated times.The relative FLIPL and FLIPS mRNA expression for each time was plotted. Resultsare representative of three independent experiments. (c) MCF10A cells incubated inthe presence or absence of EGF for the indicated times were treated with TRAIL(500 ng/ml) for 6 h and apoptosis was then determined as described in Materialsand Methods. Error bars represent S.D. from three independent experiments.**Po0.01, ***Po0.001

siRNA No

ne

FL

SC

No

ne

FL

SC

+ EGF - EGF

FLIPL

GAPDH

0

20

40

60

80

100No TRAILTRAIL

siRNA

+ EGF - EGF

Ap

op

toti

c ce

lls(%

) ***

pBabe

FLIPL

GAPDH

0

20

40

60

80No TRAILTRAIL

***

pBabe Mock

Mo

ck

FLIPL

FL

IPL

Ap

op

toti

c ce

lls (

%)

No

ne

SC FL

No

ne

SC FL

Figure 3 Role of FLIPL in EGF deprivation-promoted resistance toTRAIL-induced apoptosis. (a) MCF10A cells were transfected either with a smallinterfering RNA (siRNA) oligonucleotide targeting FLIPL (FL) or a scrambled (SC)oligonucleotide for 24 h as described in Materials and Methods. Thereafter, cellswere incubated in the presence or absence of EGF for 24 h before treatmentwith soluble TRAIL (500 ng/ml) for 6 h. Apoptosis was measured as in Figure 1.Error bars represent S.D. from three independent experiments. ***Po0.001. FLIPL

knockdown was assessed by western blotting and glyceraldehyde-3-phosphatedehydrogenase (GAPDH) was used as a protein loading control. (b) MCF10A cellswere infected with retroviruses encoding control vector (pBabe Mock) or FLIPL.Following puromycin selection for stably infected cell lines, cells were treated withsoluble TRAIL (500 ng/ml) for 6h in the presence of EGF. Apoptosis was measuredas in Figure 1. Error bars represent S.D. from three independent experiments.***Po0.001. FLIPL levels were assessed by western blotting and GAPDH wasused as a protein loading control


1910


pathways in the control of FLIPL levels, we assessed theeffect of the inhibitors U0126 (MAPK pathway) andLY294002 (PI3K/Akt pathway) on the regulation of FLIPL

levels by EGF. Re-addition of either EGF-CM or EGF aloneto EGF-deprived cells induced a marked downregulation ofFLIPL expression in MCF10A cells (Supplementary FigureS2D). Downregulation of FLIPL after re-addition of CM toEGF-deprived cells was not blocked by LY294002(Figure 4a), although AKT was significantly inhibited, whichindicated that the PI3K/Akt pathway was not involved inFLIPL regulation by EGFR signaling in MCF10A cells. Incontrast to the lack of involvement of the PI3K/Akt pathway inthe regulation of FLIPL expression by EGF, the resultsdepicted in Figure 4b (upper panel) and SupplementaryFigure S1B and S1D demonstrated that the ERK1/2 pathwayplayed an important role in the downregulation of FLIPL

expression following EGFR activation in several non-tumorbreast epithelial cell lines. Thus, inhibition of the ERK1/2pathway by the MAPK/ERK kinase 1 (MEK1) inhibitor U0126blocked FLIPL downregulation induced by EGF re-additionto EGF-deprived cells. Levels of the pro-apoptotic BH3-onlyprotein BimEL levels are shown as control of a proteinnegatively regulated by the ERK1/2 pathway in these cells.26

Furthermore, EGF-induced regulation of FLIPL expression bythe ERK1/2 pathway occurred also at the mRNA level, as thepresence of U0126 during EGF re-addition to EGF-deprivedcells blocked, at least in part, FLIPL mRNA downregulation(Figure 4b, lower panel). Importantly, inhibition of the ERK1/2pathway prevented the restoration of sensitivity to TRAILupon EGF re-addition to EGF-deprived cells (Figure 4c andSupplementary Figure S1A).

To further demonstrate the role of the ERK1/2 pathwayin EGFR-mediated regulation of FLIPL expression, wegenerated MCF10A cells overexpressing a constitutivelyactive form of MEK1 (CAMEK1). As compared to mock-transfected cells, CAMEK1 cells maintained ERK1/2 phos-phorylation after 48 h in the absence of EGF (Figure 5a).Interestingly, EGF deprivation did not result in the upregula-tion of FLIPL in CAMEK1 cells, further supporting an inhibitoryrole of the ERK1/2 pathway in the regulation of FLIPL

expression. Finally, we examined whether the EGFR-mediated regulation of sensitivity to TRAIL was altered incells stably overexpressing CAMEK1. As shown in Figure 5b,CAMEK1 cells, which maintained low cellular levels of FLIPL

even in the absence of EGF in the culture medium, were

+/+ -/- -/+ -/+ +/+ -/- -/+ -/+EGF

UT LY

FLIPL

p-AKT

AKT

GAPDH

UT U0126

FLIPL

BimELpERK1/2

ERK1/2

Tubulin

0

20

40

60

80NoTRAIL TRAIL

UT U0126

EGF +/+

Ap

op

toti

c ce

lls (

%)

EGF

0

2

4

6

8

10

12

FL

IPL

mR

NA

(Fo

ld E

xpre

ssio

n)

*** **

-/- -/+ -/+

UT U0126

EGF +/+ -/- -/+ -/+

Figure 4 Inhibition of the ERK1/2 pathway prevents FLIPL downregulation andsensitivity to TRAIL induced by EGF. MCF10A cells were preincubated in theabsence of EGF for 48 h before re-addition of EGF (� /þ ) and incubated inthe presence or absence of (a) the PI3K inhibitor LY294002 (LY, 10 mM), or (b) theMEK1 inhibitor U0126 (10 mM), for 15 h. Other cultures were incubated in parallelwith (þ /þ ) or without (� /� ) EGF for the entire experimental period(preincubation/incubation). FLIPL, p-AKT, AKT, p-ERK1/2, ERK1/2 and BimELlevels were assessed by western blotting. GAPDH and tubulin levels were used asprotein loading controls. FLIPL mRNA levels were measured by quantitative PCR(b, lower panel). Results are representative of two independent experiments.(c) Cells were preincubated/incubated as in (b) and then TRAIL (500 ng/ml) wasadded to some cultures for 6 h. Apoptosis was determined as described in Materialsand Methods. Error bars represent S.D. from three independent experiments.**Po0.01, ***Po0.001

FLIPL

BimEL

MEK1

GAPDH

+ + -- EGF

Mock CAMEK-1

p-ERK1/2

ERK1/2

0

20

40

60

80

100 NoTRAILTRAIL

Ap

op

toti

c ce

lls (

%)

pBabe Mock CAMEK-1 Mock CAMEK-1

EGF No EGF

*

Figure 5 Constitutive activation of the ERK1/2 pathway abrogates EGF-dependent regulation of FLIPL levels and sensitivity to TRAIL. (a) MCF10A cellswere infected with retroviruses encoding control vector (pBabe Mock) or aCAMEK1. Following puromycin selection for stably infected lines, cells were grownin the presence or absence of EGF for 48 h, and phosphorylated ERK1/2, totalERK1/2, FLIPL, BimEL and MEK1 levels were assessed by western blotting.GAPDH was used as a protein loading control. (b) Infected cells were incubated withor without EGF for 48 h before treatment with TRAIL (500 ng/ml) for 6 h. Thereafter,apoptosis was assessed as described in Materials and Methods. Error barsrepresent S.D. from three independent experiments. *Po0.05


1911


sensitive to TRAIL-induced apoptosis independently of EGFRactivation.

Role of c-myc in the control of FLIPL levels in MCF10Acells by EGF. C-myc is a member of the Myc familyof transcription factors that plays a central role in regulatingcell growth, cell-cycle progression and apoptosis.27 C-myc isregulated at both transcriptional and post-transcriptionallevels by the ERK1/2 pathway following activation of growthfactors receptors.16,28 It has also been reported that c-myccould repress FLIP expression at the transcriptional level.29

To determine the role of c-myc in EGFR-mediated down-regulation of FLIPL and sensitization to TRAIL-inducedapoptosis, we first examined the regulation by EGFof c-myc protein levels in MCF10A cells. Results shown inFigure 6a indicate that EGF deprivation resulted in theinhibition of the ERK1/2 pathway and c-myc downregulation.Moreover, re-addition of EGF to EGF-deprived cells triggeredthe activation of the ERK1/2 pathway and the upregulation ofc-myc in an ERK1/2-dependent manner, both events follow-ing an opposite pattern to FLIPL levels. To further investigatethe role of c-myc in the regulation of FLIPL levels by EGF, weexamined whether silencing c-myc expression could have animpact on FLIPL levels in cells maintained in EGF-CM.Results shown in Figure 6b indicate that c-myc knockdowninduced the expression of both FLIPL mRNA (left panel) andprotein (right panel) in cells grown in the presence of EGF.

As a complementary approach to further demonstrate therole of c-myc in controlling FLIPL expression in EGF-deprivedcells, we generated MCF10A cells constitutively expressing

an Myc-ER chimera.30 In these cells, activity of the chimericmyc protein is completely dependent on the addition ofexogenous 4-hydroxytamoxifen (4HT).31 As a positive controlof Myc-ER activation, we determined E2F-1 expression, atranscription factor positively regulated by c-myc,32 in cellstreated with 4HT. As shown in Figure 6c, E2F-1 wasupregulated following Myc-ER activation by 4HT in EGF-deprived cells. Interestingly, Myc-ER activation resulted inFLIPL downregulation in MCF10A cells deprived of EGF,at both the protein and mRNA levels (Figure 6c). Finally,we analyzed the sensitivity to TRAIL-induced apoptosis incells in which Myc-ER was activated by 4HT. Results shown inFigure 6d demonstrate that Myc-ER activation was sufficientto sensitize EGF-deprived cells to TRAIL.

Transformation blocks FLIPL upregulation and TRAILresistance upon EGF deprivation in MCF10A cells: roleof the ERK1/2 pathway. To investigate the impact of celltransformation on the regulation of FLIPL expression andTRAIL sensitivity by EGF, we generated MCF10A cells thatexpressed a constitutively active form of the oncogeneERBB2/HER-2/neu.33 In cells overexpressing the ERBB2(NeuT) protein, the ERK1/2 pathway was not inhibited uponEGF withdrawal (Figure 7a). Consistent with this finding,c-myc and FLIPL levels did not change in response to EGFdeprivation (Figure 7a). We next examined the regulationby EGF of TRAIL-induced apoptosis in both mock andNeuT-transduced cell lines. In contrast to what was observedin mock-transduced cells, EGF deprivation did not reduce thesensitivity to TRAIL in NeuT cells (Figure 7b), supporting the

c-myc

FLIPL

GAPDH

siRNA Scr c-myc

FL

IPL m

RN

A(f

old

exp

ress

ion

)

0

1

2

3

siRNA Scr c-myc

GAPDH

myc-ER

FLIPL

E2F-1

- + --

+EGF4HT

+

0

2

4

6

8

4HT - + -

+

+

+ - -- + - +

EGF No EGF

0

20

40

60

80

100

EGF4HT

ControlTRAIL*

Ap

op

toti

c ce

lls (

%)

FLIPL

Myc

BimEL

GAPDH

p-ERK1/2

ERK 1/2

PreincubationIncubation EGF

EGFEGF EGF/U0126-

-- -

FL

IPL

mR

NA

(fo

ld e

xpre

ssio

n)

Figure 6 Role of c-myc in the control of FLIPL levels and TRAIL sensitivity by EGF. (a) MCF10A cells were cultured in the presence or absence of EGF during 48 h, andthen EGF was added for 24 h to EGF-deprived cells. P-ERK1/2, ERK1/2, c-myc and FLIPL levels (whole cell extracts) were analyzed by western blotting. GAPDH was used asa protein loading control. Results are representative of three independent experiments. (b) MCF10A cells were transfected either with a small interfering RNA (siRNA)oligonucleotide targeting c-myc or a scrambled oligonucleotide (Scr) for 48 h as described in Materials and Methods. FLIPL levels were assessed by quantitative PCR orwestern blotting. C-myc knockdown was analyzed by western blotting. Results are representative of three independent experiments. GAPDH was used as a protein loadingcontrol. (c) Cells stably expressing a c-Myc-ER chimera were cultured in the presence or absence of EGF for 48 h before incubation in the presence or absence of 4HT(200 nM) for 30 h. Upper panel shows the protein levels of FLIPL, E2F-1 and c-Myc-ER as measured by western blotting. GAPDH was used as a protein loading control. Lowerpanel shows the analysis by quantitative PCR of FLIPL mRNA levels. Results are representative of three independent experiments. (d) c-Myc-ER cells treated as in (c) wereincubated for 6 h with TRAIL (500 ng/ml) and apoptosis was determined. Error bars represent S.D. from three independent experiments. *Po0.05


1912


hypothesis that the constitutive activation of the ERK1/2pathway and c-myc expression in these cells are responsiblefor maintaining low FLIPL levels and sensitivity to TRAIL,independently of EGF supply. Furthermore, incubation ofNeuT cells with the MEK1 inhibitor U0126 caused a markedinhibition of ERK1/2 phosphorylation and downregulation ofc-myc expression (Figure 7c). Strikingly, U0126 treatmentclearly upregulated FLIPL expression (Figure 7c) andreversed transformation-mediated sensitivity to TRAIL(Figure 7d), independently of the presence of EGF in theculture medium.

Discussion

Preclinical studies conducted in the past years have demon-strated that TRAIL as well as agonistic antibodies againstTRAIL-R1 and TRAIL-R2 can efficiently induce apoptosis intumor cells, but not in the majority of normal cells.34,35 In manyof these preclinical studies, TRAIL-sensitive cell lines wereused and it has since become clear that some primary tumorcells are TRAIL resistant.36 Moreover, it has been reportedthat EGFR pathway activation limits TRAIL-induced apoptosisin breast tumor cells19 and the small-molecule EGFR inhibitorgefitinib reverses TRAIL resistance in other tumor cells.20 Inthis respect, there are ongoing clinical trials using combinationof EGFR inhibitors and recombinant TRAIL.37 However, it israther unknown the effect that these combination treatments

may have in non-tumor cells. In our study, we demonstrate forthe first time that preventing EGFR signaling inducesresistance to TRAIL in non-tumor breast epithelial cellsthrough the upregulation of FLIPL and inhibition of TRAIL-induced apical caspase-8 activation. Our recently publisheddata demonstrate that FLIPL knockdown activates apoptosisby endogenous TRAIL in a DISC-dependent manner in thepresence of EGF and reduces the number of acini in 3Dbasement-membrane cultures of MCF10A cells.25 Conver-sely, FLIPL overexpression in MCF10A cells delays lumenformation in these cultures.25 In this respect, it has beenreported that TRAIL is upregulated during morphogenesis ofMCF10A mammary epithelial cells in 3D cultures andcooperates with Bcl-2 family proteins to regulate lumenformation in MCF10A acini.38 During morphogenesis, thecells in the center of the epithelial mass lack matrixattachment, which leads to inhibition of EGFR signaling.26 Inaccordance with our results, it would be predicted that loss ofEGFR signaling in the inner cells of the acini would induceFLIPL expression and inhibit TRAIL-induced cell death.Collectively, these results suggest that maintenance ofelevated FLIPL levels may have an important role in thetimely control of morphogenesis in the mammary gland bypreventing the early activation of TRAIL-induced cell death.

Of the signaling pathways activated upon engagement oftyrosine kinase receptors by their ligands, two of them, thePI3K/Akt and the MAPK/ERK pathways, are frequentlyreported to play a survival role preventing the inductionof apoptosis.19,39 Furthermore, inhibition of PI3K/Akt40 orMAPK/ERK41 increases the sensitivity of tumor cells toTRAIL-induced apoptosis. In contrast, other studies haveshown that sensitivity to TRAIL can be enhanced uponactivation of the PI3K/Akt42 or ERK1/243 pathways, althoughthe underlying mechanisms have not been characterized. Wehereby report for the first time that in non-tumor human breastepithelial cells, activation of the ERK1/2 MAPK pathway byEGF is crucially involved in the sensitization of these cells toTRAIL-induced apoptosis. Most of the studies examining theregulation of TRAIL sensitivity by the PI3K/Akt and MAPK/ERK pathways have been carried out in tumor cell linesof different sources in which these signaling pathwaysare commonly deregulated, leading to aberrant activation ofgenes involved in cell survival, metabolism, proliferation andmigration.44 Moreover, it is well known that the number ofactive cell surface receptors for growth factors can determinewhether ERK activation is transient or sustained, whichresults in markedly different cellular responses.45 Therefore,differences in cell surface EGFR levels may explain thedifferent outcome of EGFR activation on FLIP levels andapoptosis sensitivity between normal and tumor cells.

Our results are in agreement with previous data indicatingthat sensitivity of EGF-dependent non-tumor breast epithelialcells to different apoptotic stimuli, including death receptorligands, is abrogated in three-dimensional cultures (3D).46

In these 3D cultures, formation of a polarized structure isachieved and activation of the MAPK/ERK pathway by EGF ismarkedly attenuated.46,47 Whether or not FLIPL levelsincrease in 3D cultures need to be investigated. Interestingly,non-polarized malignant breast epithelial cells maintainan elevated MAPK/ERK pathway activation in response to

0

20

40

60

80

100No TRAILTRAIL No TRAIL

TRAIL

pBabe Mock UT U0126 UT U0126NeuT Mock NeuT

+ EGF - EGF + EGF - EGF

Ap

op

toti

c ce

lls (

%)

*

+ - + -

Mock NeuT

+ - + - EGF

FLIPL

BimEL

p-ERK 1/2

ERK 1/2

c-myc

ErbB2

GAPDH

0

20

40

60

80

100

**

FLIPL

p-ERK 1/2

c-myc

GAPDH

GAPDH

U0126

EGF

UT

BimEL

Ap

op

toti

c ce

lls (

%)

Figure 7 Oncogenic transformation blocks FLIPL upregulation and TRAILresistance upon EGF deprivation in MCF10A cells. (a) MCF10A cells were infectedeither with a retroviral control vector (Mock) or a retrovirus encoding a constitutivelyactive form of ERBB2 (NeuT). Following puromycin selection, cells were grown inthe presence or absence of EGF during 48 h and FLIPL, BimEL, ERK1/2, c-myc,ERBB2 and GAPDH levels and phospho-ERK1/2 status were analyzed by westernblotting. Results are representative of three different experiments. (b) Infected cellswere incubated as in (a) before adding TRAIL (500 ng/ml) for 6 h. Apoptosis wasmeasured as described in Materials and Methods. Error bars represent S.D. fromthree independent experiments. *Po0.05. (c) NeuT cells were incubated in thepresence or absence of EGF for 32 h. Following this incubation, the MEK1 inhibitorU0126 (10mM) was added to some cultures and the cells were incubated for 16 h.Following this incubation, cells were either collected and protein levels analyzed bywestern blotting (c) or treated with TRAIL for 6 h to assess apoptosis (d). Error barsrepresent S.D. from three independent experiments. **Po0.01


1913


EGF47 and remain sensitive to death receptor ligands in3D cultures.46

Through the phosphorylation of both cytoplasmic signalingproteins and transcription factors, activated ERK1/2 regulatescell metabolism, survival, differentiation, proliferation andmigration. Among the transcription factors regulated by ERK,the ternary complex factor plays an important role in inducingthe expression of immediate-early genes such as c-Fos andc-Myc.48 In addition, activation of the ERK1/2 pathwayinduces stabilization of the c-myc protein.49 Interestingly,our data demonstrate that EGF-dependent activation of theERK1/2 pathway enhances c-Myc levels and negativelyregulates FLIPL expression in breast epithelial cells, andsensitizes these cells to TRAIL-induced apoptosis. By usingeither siRNA oligonucleotides to downregulate c-myc or theestrogen activatable Myc-ER chimera, we have providedfurther evidences supporting that c-myc plays a regulatoryrole in the control of FLIPL levels by EGF, which are inagreement with previous data indicating that c-myc binds andrepresses the human FLIP promoter and sensitizes cells toTRAIL.29 However, in tumor cells c-myc expression isfrequently dissociated from the ERK1/2 pathway, either bymutation, reduced degradation through the ubiquitin–proteasome system or by other mechanisms.49,50 Accord-ingly, in tumor cells FLIPL levels could be regulated in anERK1/2-independent manner by other survival pathways.51

Our results also indicate that in EGF-deprived cells, c-mycactivation using a 4HT-activatable Myc-ER chimera doesnot completely reproduce the sensitization achieved by EGFre-addition, which suggests that other effectors downstreamof ERK1/2 activation could be also important for a completeapoptotic response to TRAIL. In this respect, it has beenreported that c-Fos can function as a proapoptotic protein byrepressing the expression of FLIPL through binding to thegene promoter.52 Alternatively, other signaling emerging fromthe EGFR such as the p38-MAPK and the Jun N-terminalkinase (JNK) pathways might downregulate FLIPL expressionby transcriptional and post-transcriptional mechanisms. Thus,it has been demonstrated that EGFR-mediated activation ofthe p38 pathway regulates the activity of the transcriptionfactor SP3,53 which under certain conditions could be anegative regulator of the FLIPL gene promoter.54 Moreover,JNK activation by TNF-a reduces FLIPL stability by amechanism involving the JNK-mediated phosphorylationand activation of the E3 ubiquitin ligase Itch, which ubiquiti-nates FLIPL and induces its proteasomal degradation.55

Overexpression of oncogenic receptor tyrosine kinases is acommon event in breast cancer. In particular, 15–30% of allcases show elevated ERBB2,56 but despite the developmentof ERBB2/HER2-targeted therapies, only 35% of ERBB2-positive patients initially respond to those treatments. Ithas been shown in experiments in vitro40 and in vivo57 thatcombination of antibodies against ERBB2 and TRAIL recep-tors facilitates apoptosis and tumor regression, although thereare data reporting that the apoptosis-inducing capacity ofthese combinations is cell type-dependent.58 Our resultsindicate that in ERBB2-overexpressing cells sensitivity toTRAIL is controlled by the ERK1/2 pathway-mediated regula-tion of FLIPL levels. These data suggest that amplification ofERBB2 in tumor cells may have different outcomes regarding

sensitivity to TRAIL. On one hand, it may increase resistanceto TRAIL through ERBB2-induced activation of the PI3K/Aktpathway.40 On the other hand, it may contribute to maintainlow FLIPL levels by ERK1/2-mediated activation of c-myc andother genes,29,52 which may result in enhanced sensitivity toTRAIL. This is not unique of ERBB2 as other oncoproteinscould also sensitize cells to TRAIL by activating the ERK1/2pathway,59 although the mechanism underlying this sensitiza-tion has not been elucidated. Our data highlight the role ofthe EGF-regulated, ERK1/2 pathway-mediated regulation ofFLIPL levels as an important mechanism modulating thesensitivity of human breast epithelial cells to TRAIL-inducedapoptosis that may contribute, in concert with others, to thedifferential sensitivity of normal and tumor cells to TRAIL.At the same time, our results provide arguments for a cautiousclinical application of TRAIL in cancer patients, especially incombination with agents that may inhibit the ERK1/2 pathway.

Materials and MethodsReagents and antibodies. Recombinant human EGF was from Peprotech(London, UK). Recombinant human TRAIL (residues 95–281) was produced asdescribed previously.60 U0126 and gefitinib were purchased from SelleckChemicals (Houston, TX, USA). Mouse anti-a-tubulin antibody, LY294002, 4HT,hydrocortisone, transferrin and puromycin were obtained from Sigma-Aldrich(St. Louis, MO, USA). Anti-caspase-8 was generously provided by Dr. GeraldCohen (Leicester University, Leicester, UK). Anti-FADD, anti-ERBB2 and anti-E2F1 monoclonal antibodies were obtained from BD Biosciences (Erembodegem,Belgium). GAPDH and c-myc monoclonal antibodies were from Santa CruzTechnology (Santa Cruz, CA, USA). Anti-TRAIL-R2 and anti-c-FLIP monoclonalantibody (NF6) were from Alexis Corporation (Lausen, Switzerland). Anti-TRAIL-R1 and anti-TRAIL-R2 monoclonal antibodies for surface receptor analysis werefrom Abcam (Cambridge, UK). Anti-pAKT, anti-AKT, anti-pERK1/2 and anti-MEK1antibodies were obtained from Cell Signaling Technology (Temecula, CA, USA).Anti-ERK antibody was from Upstate-Millipore (New York, NY, USA). Anti-Bimpolyclonal antibody was purchased from Calbiochem (Darmstadt, Germany).Horseradish peroxidase or FITC-conjugated secondary antibodies, goat anti-mouse and goat anti-rabbit were obtained from DAKO (Cambridge, UK).

Cell lines. MCF10A and MCF12A cell lines were maintained in DMEM/F12supplemented with 5% donor horse serum, 2 mM L-glutamine, 20 ng of EGF perml, 10mg of insulin per ml, 100 ng of cholera toxin per ml, 0.5mg of hydrocortisoneper ml, 50 U of penicillin per ml and 50mg of streptomycin per ml at 37 1C in a 5%CO2-humidified, 95% air incubator. The 184A1 cells were cultured in the samemedium with transferrin (5 mg/ml).

Determination of apoptosis. Cells (3� 105 per well) were treated in6-well plates as indicated in the figure legends. After treatment, hypodiploidapoptotic cells were detected by flow cytometry according to publishedprocedures.60 Basically, cells were washed with phosphate-buffered saline(PBS), fixed in cold 70% ethanol and then stained with propidium iodide whiletreating with RNAse. Quantitative analysis of sub-G1 cells was carried out in aFACSCalibur cytometer using the Cell Quest software (Becton Dickinson,Mountain View, CA, USA).

Immunoblot analysis of proteins. Cells (3� 105) were washed withPBS and protein content was measured following cell lysis using the Bradfordreagent (Bio-Rad Laboratories, Hercules, CA, USA) before adding Laemmlisample buffer. Where indicated, whole cell extracts were obtained. Sampleswere sonicated, and proteins were resolved on SDS-polyacrylamide minigels anddetected as described previously.60

Analysis of TRAIL receptors by flow cytometry. MCF10A cells weredetached with RPMI/3 mM EDTA, and cytofluorimetric analysis of proteins wasperformed as described previously.60 Briefly, cells were washed in ice-cold PBS,and resuspended in PBS. Cells were then labeled with anti-TRAIL receptorantibodies (5 mg/ml) or no antibody (negative control) and then incubated with goat


1914


anti-mouse FITC-conjugated antibody (F(ab)2 fragment). Labeled cells wereanalyzed by flow cytometry using the Cell Quest software.

Real-time RT-PCR. Total RNA was isolated from MCF10A cells with theTrizol reagent (Life Technologies, Grand Island, NY, USA) as recommended bythe supplier. Total RNA (2 mg) was used as a template for cDNA synthesis usingan RT-PCR kit (Perkin-Elmer, Waltham, MA, USA). mRNA expression wasanalyzed in triplicate by quantitative RT-PCR on the ABI Prism7500 sequencedetection system using predesigned Assay-on-demand primers and probes(Applied Biosystems, Carlsbad, CA, USA). mRNA expression was determined bythe comparative cycle threshold (Ct) method (DDCt). Hypoxanthine-guaninephosphoribosyltransferase (HPRT Part Number 4331182 and HPRT1 PartNumber 4351370) was used as an internal control and mRNA expression levelsof FLIPL and cFLIPS were given as fraction of mRNA levels in control cells.Primers and probes used were: cFLIPS forward, 50-CAGCAATCCAAAAGAGTCT-CAAGGA-30; cFLIPS reverse, 50-AATTTTCAGATCAGGACAATGGGCATA-30;probe FLIPS: 50-ACTTCAGGATGATAACACCC-30; for FLIPL we used either oftwo different oligos and probes: FLIPL forward, 50-AGTGCCTCTCCCA-GAAACTGA-30; FLIPL reverse, 50-GCTGTTCCAATCATACATGTAGCCATT-30;probe FLIPL, 50-CAAGAAAGAAAACGCCCACTCC-30; FLIPL 2 forward, 50-GGCTCCCCCTGCATCAC-30; FLIPL 2 reverse, 50-TTTGGCTTCCCTGCTAGATAAGG-30; probe FLIPL 2, 50-CAGGAGGATGTTCATGGGAGATTCATGC-30.

RNA interference. siRNAs against FLIPL: 50-CCUAGGAAUCUGCCUGAUAdTdT-30 and non-targeting scrambled siRNA were synthesized by SigmaProligo (St. Louis, MO, USA). siRNA against c-Myc was purchased from Qiagen(Germantown, PA, USA) (FlexiTube). Cells were transfected with 50 nM siRNAsusing DharmaFECT-1 (Dharmacon, Lafayette, CA, USA) as described by themanufacturer. After 24 h for FLIPL siRNA or 48 h for c-myc siRNA, transfectionmedium was replaced with regular medium before further analysis.

Retroviral vectors and virus production. FLIPL (in pCR3.V64 vector, akind donation of Dr. J Tschopp, University of Lausanne) was cloned into BamHI/SalIsites of pBabepuro. pBabepuro constitutively active MEK (S217E/S221E) was a giftfrom Dr. CJ Marshall (Institute of Cancer Research, London, UK). pBabepuro-myc-ER was donated by Dr. J Leon (Instituto de Biomedicina y Biotecnologıa deCantabria, Santander, Spain). Constitutively active ErbB2 mutant (pBabe-NeuT) waskindly provided by Danielle Carroll (Harvard Medical School, Boston, MA, USA).Retroviruses for protein overexpression were produced by transfection of HEK293-Tcells by calcium phosphate method with the corresponding retroviral vectors.Retrovirus-containing supernatants were collected 48 h after transfection andconcentrated by ultracentrifugation at 22 000 r.p.m. for 90 min at 41C.

Generation of MCF10A cell lines. MCF10A cells were plated at 3.5� 105

cells per 10-cm dish and infected with the retroviruses mentioned above. Stablepopulations were obtained by selection with 1.5mg/ml puromycin during 48 h.

Statistical Analysis. All data are presented as the mean±S.D. of at leastthree independent experiments. The differences among different groups weredetermined by the Student’s t-test. Po0.05 was considered significant.

Conflict of InterestThe authors declare no conflict of intrest.

Acknowledgements. We thank Dr. Gemma Fuster and Dr. Christophe Vandierfor kindly providing us with the MCF12A and 184A1 cell lines, respectively. This workwas supported by grants from Ministerio de Ciencia e Innovacion (SAF2009-07163),Red Tematica de Investigacion Cooperativa en Cancer (RTICC: RD06/0020/0068), theEuropean Community through the regional development funding program (FEDER)and Junta de Andalucıa (P09-CVI-4497) to ALR. CP and RY were supported bycontracts from Ministerio de Ciencia e Innovacion and Junta de Andalucıa, respectively.

1. Wiley SR, Schooley K, Smolak PJ, Din WS, Huang CP, Nicholl JK et al. Identificationand characterization of a new member of the TNF family that induces apoptosis.Immunity 1995; 3: 673–682.

2. Johnstone RW, Frew AJ, Smyth MJ. The TRAIL apoptotic pathway in cancer onset,progression and therapy. Nat Rev Cancer 2008; 8: 782–798.

3. Bodmer JL, Holler N, Reynard S, Vinciguerra P, Schneider P, Juo P et al. TRAIL receptor-2signals apoptosis through FADD and caspase-8. Nat Cell Biol 2000; 2: 241–243.

4. Sprick MR, Weigand MA, Rieser E, Rauch CT, Juo P, Blenis J et al. FADD/MORT1 andcaspase-8 are recruited to TRAIL receptors 1 and 2 and are essential for apoptosismediated by TRAIL receptor 2. Immunity 2000; 12: 599–609.

5. Irmler M, Thome M, Hahne M, Schneider P, Hofmann K, Steiner V et al. Inhibition of deathreceptor signals by cellular FLIP. Nature 1997; 388: 190–195.

6. Chang DW, Xing Z, Pan Y, Algeciras-Schimnich A, Barnhart BC, Yaish-Ohad S et al.c-FLIP(L) is a dual function regulator for caspase-8 activation and CD95-mediatedapoptosis. EMBO J 2002; 21: 3704–3714.

7. Krueger A, Schmitz I, Baumann S, Krammer PH, Kirchhoff S. Cellular FLICE-inhibitoryprotein splice variants inhibit different steps of caspase-8 activation at the CD95 death-inducing signaling complex. J Biol Chem 2001; 276: 20633–20640.

8. Pop C, Oberst A, Drag M, Van Raam BJ, Riedl SJ, Green DR et al. FLIP(L) inducescaspase 8 activity in the absence of interdomain caspase 8 cleavage and alters substratespecificity. Biochem J 2011; 433: 447–457.

9. Kreuz S, Siegmund D, Scheurich P, Wajant H. NF-kappaB inducers upregulatecFLIP, a cycloheximide-sensitive inhibitor of death receptor signaling. Mol Cell Biol2001; 21: 3964–3973.

10. Fukazawa T, Fujiwara T, Uno F, Teraishi F, Kadowaki Y, Itoshima T et al. Accelerateddegradation of cellular FLIP protein through the ubiquitin–proteasome pathway inp53-mediated apoptosis of human cancer cells. Oncogene 2001; 20: 5225–5231.

11. Citri A, Yarden Y. EGF-ERBB signalling: towards the systems level. Nat Rev Mol Cell Biol2006; 7: 505–516.

12. Meloche S, Pouyssegur J. The ERK1/2 mitogen-activated protein kinase pathway as amaster regulator of the G1- to S-phase transition. Oncogene 2007; 26: 3227–3239.

13. Gille H, Kortenjann M, Thomae O, Moomaw C, Slaughter C, Cobb MH et al. ERKphosphorylation potentiates Elk-1-mediated ternary complex formation and transactivation.EMBO J 1995; 14: 951–962.

14. Murphy LO, Smith S, Chen RH, Fingar DC, Blenis J. Molecular interpretation of ERK signalduration by immediate early gene products. Nat Cell Biol 2002; 4: 556–564.

15. Okazaki K, Sagata N. The Mos/MAP kinase pathway stabilizes c-Fos by phosphorylationand augments its transforming activity in NIH 3T3 cells. EMBO J 1995; 14: 5048–5059.

16. Sears R, Leone G, DeGregori J, Nevins JR. Ras enhances Myc protein stability.Mol Cell 1999; 3: 169–179.

17. Normanno N, De Luca A, Bianco C, Strizzi L, Mancino M, Maiello MR et al. Epidermalgrowth factor receptor (EGFR) signaling in cancer. Gene 2006; 366: 2–16.

18. Baselga J, Arteaga CL. Critical update and emerging trends in epidermal growth factorreceptor targeting in cancer. J Clin Oncol 2005; 23: 2445–2459.

19. Gibson EM, Henson ES, Haney N, Villanueva J, Gibson SB. Epidermal growthfactor protects epithelial-derived cells from tumor necrosis factor-related apoptosis-inducing ligand-induced apoptosis by inhibiting cytochrome c release. Cancer Res 2002;62: 488–496.

20. Shrader M, Pino MS, Lashinger L, Bar-Eli M, Adam L, Dinney CP et al. Gefitinib reversesTRAIL resistance in human bladder cancer cell lines via inhibition of AKT-mediatedX-linked inhibitor of apoptosis protein expression. Cancer Res 2007; 67: 1430–1435.

21. Barker AJ, Gibson KH, Grundy W, Godfrey AA, Barlow JJ, Healy MP et al. Studies leadingto the identification of ZD1839 (IRESSA): an orally active, selective epidermal growth factorreceptor tyrosine kinase inhibitor targeted to the treatment of cancer. Bioorg Med ChemLett 2001; 11: 1911–1914.

22. Soule HD, Maloney TM, Wolman SR, Peterson WD Jr, Brenz R, McGrath CM et al.Isolation and characterization of a spontaneously immortalized human breast epithelial cellline, MCF-10. Cancer Res 1990; 50: 6075–6086.

23. Shankaran H, Ippolito DL, Chrisler WB, Resat H, Bollinger N, Opresko LK et al. Rapid andsustained nuclear–cytoplasmic ERK oscillations induced by epidermal growth factor.Mol Syst Biol 2009; 5: 332–344.

24. Palacios C, Yerbes R, Lopez-Rivas A. Flavopiridol induces cellular FLICE-inhibitory proteindegradation by the proteasome and promotes TRAIL-induced early signaling andapoptosis in breast tumor cells. Cancer Res 2006; 66: 8858–8869.

25. Yerbes R, Palacios C, Reginato MJ, Lopez-Rivas A. Cellular FLIP(L) plays a survival roleand regulates morphogenesis in breast epithelial cells. Biochim Biophys Acta 2011; 1813:168–178.

26. Reginato MJ, Mills KR, Paulus JK, Lynch DK, Sgroi DC, Debnath J et al. Integrins andEGFR coordinately regulate the pro-apoptotic protein Bim to prevent anoikis. Nat Cell Biol2003; 5: 733–740.

27. Eilers M, Eisenman RN. Myc’s broad reach. Genes Dev 2008; 22: 2755–2766.28. Cheng M, Wang D, Roussel MF. Expression of c-Myc in response to colony-stimulating

factor-1 requires mitogen-activated protein kinase kinase-1. J Biol Chem 1999; 274:6553–6558.

29. Ricci MS, Jin Z, Dews M, Yu D, Thomas-Tikhonenko A, Dicker DT et al. Direct repression ofFLIP expression by c-myc is a major determinant of TRAIL sensitivity. Mol Cell Biol 2004;24: 8541–8555.

30. Eilers M, Picard D, Yamamoto KR, Bishop JM. Chimaeras of myc oncoprotein and steroidreceptors cause hormone-dependent transformation of cells. Nature 1989; 340: 66–68.

31. Littlewood TD, Hancock DC, Danielian PS, Parker MG, Evan GI. A modified oestrogenreceptor ligand-binding domain as an improved switch for the regulation of heterologousproteins. Nucleic Acids Res 1995; 23: 1686–1690.


1915


32. Leone G, DeGregori J, Sears R, Jakoi L, Nevins JR. Myc and Ras collaborate in inducingaccumulation of active cyclin E/Cdk2 and E2F. Nature 1997; 387: 422–426.

33. Haenssen KK, Caldwell SA, Shahriari KS, Jackson SR, Whelan KA, Klein-Szanto AJ et al.ErbB2 requires integrin alpha5 for anoikis resistance via Src regulation of receptor activityin human mammary epithelial cells. J Cell Sci 2010; 123: 1373–1382.

34. Ashkenazi A, Dixit VM. Apoptosis control by death and decoy receptors. Curr Opin Cell Biol1999; 11: 255–260.

35. Walczak H, Miller RE, Ariail K, Gliniak B, Griffith TS, Kubin M et al. Tumoricidal activity oftumor necrosis factor-related apoptosis-inducing ligand in vivo. Nat Med 1999; 5: 157–163.

36. Grosse-Wilde A, Voloshanenko O, Bailey SL, Longton GM, Schaefer U, Csernok AI et al.TRAIL-R deficiency in mice enhances lymph node metastasis without affecting primarytumor development. J Clin Invest 2008; 118: 100–110.

37. Yee L, Burris H, Kozloff M et al. Phase 1b study of recombinant human APO2L/TRAIL plusirinotecan and cetuximab or FOLFIRI in metastatic colorectal cancer (mCRC) patients:preliminary results. J Clin Oncol (ASCO Meet Abstr) 2009; 27: 4129.

38. Mills KR, Reginato M, Debnath J, Queenan B, Brugge JS. Tumor necrosis factor-relatedapoptosis-inducing ligand (TRAIL) is required for induction of autophagy during lumenformation in vitro. Proc Natl Acad Sci USA 2004; 101: 3438–3443.

39. Bonni A, Brunet A, West AE, Datta SR, Takasu MA, Greenberg ME. Cell survival promotedby the Ras-MAPK signaling pathway by transcription-dependent and -independentmechanisms. Science 1999; 286: 1358–1362.

40. Cuello M, Ettenberg SA, Clark AS, Keane MM, Posner RH, Nau MM et al. Down-regulationof the erbB-2 receptor by trastuzumab (herceptin) enhances tumor necrosis factor-relatedapoptosis-inducing ligand-mediated apoptosis in breast and ovarian cancer cell lines thatoverexpress erbB-2. Cancer Res 2001; 61: 4892–4900.

41. Zhang XD, Borrow JM, Zhang XY, Nguyen T, Hersey P. Activation of ERK1/2 protectsmelanoma cells from TRAIL-induced apoptosis by inhibiting Smac/DIABLO release frommitochondria. Oncogene 2003; 22: 2869–2881.

42. Kadowaki Y, Chari NS, Teo AE, Hashi A, Spurgers KB, McDonnell TJ. PI3 kinase inhibitionon TRAIL-induced apoptosis correlates with androgen-sensitivity and p21 expression inprostate cancer cells. Apoptosis 2011; 16: 627–635.

43. Frese S, Pirnia F, Miescher D, Krajewski S, Borner MM, Reed JC et al. PG490-mediatedsensitization of lung cancer cells to Apo2L/TRAIL-induced apoptosis requires activation ofERK2. Oncogene 2003; 22: 5427–5435.

44. Shaw RJ, LC. Cantley LC. RasPI(3)K and mTOR signalling controls tumour cell growth.Nature 2006; 441: 424–430.

45. Marshall CJ. Specificity of receptor tyrosine kinase signaling: transient versus sustainedextracellular signal-regulated kinase activation. Cell 1995; 80: 179–185.

46. Weaver VM, Lelievre S, Lakins JN, Chrenek MA, Jones JC, Giancotti F et al. beta4 integrin-dependent formation of polarized three-dimensional architecture confers resistance toapoptosis in normal and malignant mammary epithelium. Cancer Cell 2002; 2: 205–216.

47. Wang F, Weaver VM, Petersen OW, Larabell CA, Dedhar S, Briand P et al.Reciprocal interactions between beta1-integrin and epidermal growth factor receptor inthree-dimensional basement membrane breast cultures: a different perspective in epithelialbiology. Proc Natl Acad Sci USA 1998; 95: 14821–14826.

48. Mendoza MC, Er EE, Blenis J. The Ras-ERK and PI3K-mTOR pathways: cross-talk andcompensation. Trends Biochem Sci 2011; 36: 320–328.

49. Sears R, Nuckolls F, Haura E, Taya Y, Tamai K, Nevins JR. Multiple Ras-dependentphosphorylation pathways regulate Myc protein stability. Genes Dev 2000; 14: 2501–2514.

50. Yada M, Hatakeyama S, Kamura T, Larabell CA, Dedhar S, Briand P et al.Phosphorylation-dependent degradation of c-Myc is mediated by the F-box proteinFbw7. EMBO J 2004; 23: 2116–2125.

51. Gibson SB. Epidermal growth factor and trail interactions in epithelial-derived cells. VitamHorm 2004; 67: 207–227.

52. Zhang X, Zhang L, Yang H, Huang X, Otu H, Libermann TA et al. c-Fos as a proapoptoticagent in TRAIL-induced apoptosis in prostate cancer cells. Cancer Res 2007; 67:9425–9434.

53. Xu K, Shu HK. EGFR activation results in enhanced cyclooxygenase-2 expression throughp38 mitogen-activated protein kinase-dependent activation of the Sp1/Sp3 transcriptionfactors in human gliomas. Cancer Res 2007; 67: 6121–6129.

54. Ganapathy M, Ghosh R, Jianping X, Zhang X, Bedolla R, Schoolfield J et al. Involvement ofFLIP in 2-methoxyestradiol-induced tumor regression in transgenic adenocarcinoma ofmouse prostate model. Clin Cancer Res 2009; 15: 1601–1611.

55. Chang L, Kamata H, Solinas G, Luo JL, Maeda S, Venuprasad K et al. The E3 ubiquitinligase itch couples JNK activation to TNFalpha-induced cell death by inducing c-FLIP(L)turnover. Cell 2006; 124: 601–613.

56. Slamon DJ, Clark GM, Wong SG, Levin WJ, Ullrich A, McGuire WL. Human breastcancer: correlation of relapse and survival with amplification of the HER-2/neu oncogene.Science 1987; 235: 177–182.

57. Stagg J, Sharkey J, Pommey S, Young R, Takeda K, Yagita H et al. Antibodies targeted toTRAIL receptor-2 and ErbB-2 synergize in vivo and induce an antitumor immune response.Proc Natl Acad Sci USA 2008; 105: 16254–16259.

58. Dubska L, Andera L, Sheard MA. HER2 signaling downregulation by trastuzumab andsuppression of the PI3K/Akt pathway: an unexpected effect on TRAIL-induced apoptosis.FEBS Lett 2005; 579: 4149–4158.

59. Nesterov A, Nikrad M, Johnson T, Kraft AS. Oncogenic Ras sensitizes normal human cellsto tumor necrosis factor-alpha-related apoptosis-inducing ligand-induced apoptosis.Cancer Res 2004; 64: 3922–3927.

60. Ruiz-Ruiz C, Lopez-Rivas A. Mitochondria-dependent and -independent mechanisms intumour necrosis factor-related apoptosis-inducing ligand (TRAIL)-induced apoptosis areboth regulated by interferon-gamma in human breast tumour cells. Biochem J 2002; 365:825–832.

Supplementary Information accompanies the paper on Cell Death and Differentiation website (http://www.nature.com/cdd)


1916


http://www.nature.com/cdd

Documents

Control of FLIPL expression and TRAIL resistance by the ...file2.selleckchem.com/citations/getinib-U0126-20120343.pdfthe extracellular signal-regulated kinase1/2 pathway in breast